Abstract 15552: Use of Risk Estimates During Percutaneous Coronary Intervention: Physician-Level Barrier of Commission Versus Omission
Background: Treatment outcomes vary according to individual patient characteristics; thus, some patients may derive greater benefit from a specific treatment than others. For example, vessel restenosis after percutaneous coronary intervention (PCI) is both predictable and modifiable. Although drug eluting stents (DES) reduce restenosis rates, recent data suggest that patients’ restenosis risk has little impact on DES use. PRISM is a tool that generates validated, individualized risk estimates to support clinical decision making, such as stent selection. The present study explored the use of risk estimates for patients undergoing PCI by interventional cardiologists’ (ICs) from 8 U.S. hospitals which recently implemented PRISM.
Methods: A multidisciplinary team conducted a qualitative descriptive study between July 2011 and Feb 2012. ICs were invited (n=57) and interviewed in-person (n=5) or by telephone (n=22); 3 were IC fellows; 2 were female. Average interventional experience was 13 (0.5 - 30) years. All interviews were recorded (mean 26 min), transcribed verbatim and then coded using descriptive content analysis until thematic saturation occurred.
Results: ICs believed PRISM was beneficial as an informational tool for patients, but not as a physician decision making tool. A predominant theme was the perception that the use of risk models leads to omission of therapy: “At the end of the day, you want to do the best thing you can do for each and every patient, not just the high risk patients” (IC#44-24; 9.5 years experience); and “The expressed goal of the restenosis model is in some way to withdraw therapies. I think physicians have a harder time withdrawing therapies than adding an effective therapy. The biggest barrier to using the restenosis [model] is I’m not aligned with the strategy ¼ to withdraw therapy” (IC#01-04; 12 years experience). Additionally, an IC stated “you want to put the Cadillac in everybody” (IC#01-03; 20 years experience).
Conclusions: The perception that risk modeling leads to omission of medical therapy is a significant barrier to the practice of personalized medicine. It is necessary to overcome this barrier so that health care policy and providers can better target medical therapies to those patients most likely to benefit.
- © 2012 by American Heart Association, Inc.